WFPC2 Instrument Science Report 96-061
Photometric Calibration of WFPC2
Linear Ramp Filter Data in SYNPHOT
J. Biretta, S. Baggett, and Keith Noll
July 10, 1996
ABSTRACT
We discuss derivation and use of the SYNPHOT Linear Ramp Filter throughput table,
which provides a preliminary photometric calibration with accuracy ~3%.
1. Introduction
The SYNPHOT synthetic photometry package has become one of the main tools for performing photometric calibration of WFPC2 data. In this report we derive a SYNPHOT filter transmission table for the Linear Ramp Filters (LRFs). This new table, together with
system and CCD efficiencies already in SYNPHOT, provides a preliminary photometric
calibration of the LRFs.
This table has two principal ingredients. The first is the TRANS LRF table, which maps
the observer’s central wavelength onto a particular spot on the LRF filter glass. (The
TRANS LRF table can be found in Biretta, et al., 1996a as Table 13; an abridged version
appears in Tables 3.7 and 3.8 of the WFPC2 Handbook, Biretta, et al., 1996b) The second
ingredient is a set of ground-based calibrations which give the LRF filter transmissions as
a function of position on the filter glass. In this report we discuss how the TRANS LRF
table and the laboratory transmission measurements are used together to generate the
SYNPHOT LRF table. Sections 2 and 3 review the general properties of the LRFs and the
lab measurements made at JPL. Section 4 describes how the table is computed. Section 5
describes how to obtain and install the new table, and Section 6 describes how to use this
table for photometric calibration.
Our experience with other WFPC2 filters indicates that the lab filter calibration, together
with on-orbit determinations of CCD and system throughput, should provide photometric
calibration accurate to ~3%, which is sufficient for most purposes. In the future, standard
1. Copies of this report may be obtained from the Science Support Division, Space Telescope Science Institute, 3700 San Martin Drive, Baltimore MD 21218, by e-mail to help@stsci.edu, by anonymous FTP to stsci.edu directory instrument_news/WFPC2, and by WWW at http://www.stsci.edu/
instruments.html
1
stars will be observed through the LRFs to verify the predicted throughputs.
All SYNPHOT users are advised to obtain a copy of the current (March 1995) SYNPHOT
User’s Guide by sending e-mail to help@stsci.edu.
Table of Contents
Section
Topic
Page
1
Introduction
1
2
Review of LRF Filter Properties
3
3
Ground-based Transmission Measurements for the LRFs
3
4
Calculation of the SYNPHOT LRF Table
7
...4.1
Format and utilization of the SYNPHOT LRF table
7
...4.2
Calculation of the SYNPHOT LRF table
9
5
Obtaining and Installing the SYNPHOT LRF Table
10
6
How to Perform LRF Photometric Calibration with SYNPHOT
12
...6.1
Emission line spectra
14
...6.2
Stellar spectra
17
...6.3
Power law spectra
19
7
References
19
Appendix
Sample LRF Transmission Curves
20
2
2. Review of LRF Filter Properties
The WFPC2 Linear Ramp Filters (LRFs) offer a narrow band imaging capability which is
tunable from 3710Å to 9762Å with bandpass FWHM ∆λ / λ ~0.013. The LRF filter set
contains four narrow band interference filters (named FR418N, FR533N, FR680N, and
FR868N), and each filter contains four ramps (numbered #1 through #4) whose central
wavelength varies as a function of position on the filter surface. In addition, each filter can
be rotated to four different partial rotations of the filter wheel (-33˚, -18˚, 0˚, or +15˚); this
rotation is used to move wavelengths which would otherwise fall outside the CCDs into
the region imaged by the CCD detectors.
An observer using the LRFs is asked only to specify the desired “central wavelength” on
the Phase II proposal; the aperture and filter is entered simply as “LRF.” During proposal
implementation, the TRANS program must convert the user-specified wavelength into a
correct filter selection, and must also calculate a positional offset that will place the target
at the correct location on the ramp filter. This is done by finding the requested central
wavelength in the TRANS LRF lookup table, and then interpolating the pointing offset
from the two nearest entries in the table. The table also specifies the filter, filter wheel rotation, and CCD to be used.
3. Ground-based Transmission Measurements for the LRFs
Wavelength and transmission calibration is provided by a set of lab measurements made
on the flight filters at JPL. A complete description of the LRF transmission measurements
can be found in Evans (1992). A monochrometer with a narrow slit was used to provide
illumination, and a micrometer provided position information. The transmission profile
was measured at five equally spaced points along the length of each ramp. The resulting
data were later integrated over an annular aperture to simulate the out-of-focus OTA beam
passing through the filter. This integration largely defines the bandpass shape, and results
in highly symmetric transmission curves.
Munson functions of the following form were then fitted to the integrated transmission
curves:
T0
T = -----------------------------------------------------------------------------------------------------------------------2
1 + ( 1 – a )x + a ( 1 – b )x 4 + ab ( 1 – c )x 6 + abcx 8
where
a, b, and c are shape parameters ( 0 ≤ a, b, c ≤ 1 ),
T0 is the peak transmission of the passband, T=T0 at x=0,
3
(1)
x is related to wavelength by
(w – w0)
x = -------------------HWHM
(2)
and w0 is the center wavelength.
The shape parameters, a, b, and c, and the parameters T0, w0, and HWHM were then fitted
to polynomial functions of physical distance Y in inches along the ramp:
Parameter = A 0 + A 1 Y + A 2 Y 2 + A 3 Y 3
(3)
Table 1 gives the coefficients in this equation. These coefficients are identical to those
listed in Evans (1992) and in WFPC2 Handbook (V.3.0) Tables 3.3 to 3.5, except for the
parameter Y. For consistency with the TRANS LRF table, we instead use the linearized
wavelength relations given by Biretta, et al. (1996a), which are typically accurate to 0.3Å:
Y = A0 + A1w0
(4)
The coefficients in equation (4) are derived from C1,2 and C1,4 given in Table 3 of Biretta,
et al. (1996a):
A 0 = C 1, 2 C 1, 4
A 1 = C 1, 4
and are listed in Table 1.
Table 1: Polynomial Coefficients for Parameters.
Filter / Ramp
#
FR418N #1
FR418N #2
Parameter
Y
w0
T0
HWHM
a
b
c
Y
w0
A0
-26.1083
3657.7
-.01667
21.95
.2120
1.181
.3301
-24.2554
3876.9
A1
.00713888
138.7
.2188
-.8347
.002857
-.8138
-.3715
.00625704
158.6
4
A2
.0000
.6178
.04138
2.143
.002596
.3535
.3825
.0000
.5472
A3
.0000
.0000
-.03489
.0000
.0000
.0000
.0000
.0000
.0000
Table 1: Polynomial Coefficients for Parameters.
Filter / Ramp
#
FR418N #3
FR418N #4
FR533N #1
FR533N #2
FR533N #3
Parameter
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
A0
.1660
21.50
.1592
.7938
.9306
-24.7145
4130.5
.1352
22.09
.2300
1.096
1.276
-23.4440
4371.3
.3189
25.62
.3123
.9222
1.033
-26.7670
4677.7
.5450
25.67
-.009839
.31511
-.3379
-24.6600
4948.4
.4546
32.10
.1678
.9345
.9571
-24.5038
5257.3
.4944
34.03
.3851
.5605
A1
.2288
3.315
-.003687
.2355
.01366
.00598254
168.8
.6200
1.306
.05586
-.3185
-1.279
.00536340
185.8
.1287
1.015
-.2055
.1167
-.1356
.00572115
177.3
-.3612
.3168
.4644
.9473
2.788
.00498393
199.2
.4188
-1.204
-.02726
.1935
.02919
.00465985
217.9
-.1714
5.078
-.06264
.6642
5
A2
-.1080
-.7079
-.0008497
-.09124
.007458
.0000
-.7389
-.5226
-.1181
-.03044
.1396
.5721
.0000
.2913
-.01160
.1161
.09535
-.04673
.05660
.0000
-1.125
.3623
.8873
-.2039
-.4516
-1.346
.0000
.6484
-.5456
3.171
.09521
-.1244
-.009393
.0000
-1.481
.1890
-1.347
.003163
-.2751
A3
.004005
.0000
.0000
.0000
.0000
.0000
.0000
.1529
.0000
.0000
.0000
.0000
.0000
.0000
-.001712
.0000
.0000
.0000
.0000
.0000
.0000
-.1281
.0000
.0000
.0000
.0000
.0000
.0000
.1548
.0000
.0000
.0000
.0000
.0000
.0000
-.0631
.0000
.0000
.0000
Table 1: Polynomial Coefficients for Parameters.
Filter / Ramp
#
FR533N #4
FR680N #1
FR680N #2
FR680N #3
FR680N #4
FR868N #1
Parameter
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
A0
.9665
-25.5182
5596.9
.5058
35.06
.06553
1.043
1.162
-21.8962
5916.0
.1198
41.50
.1743
.8320
.9682
-22.6919
6290.8
.7918
39.48
.05596
1.017
.7244
-22.0719
6673.5
.9494
42.81
.1038
.8415
1.017
-24.7447
7141.9
.4823
44.72
.1612
.2708
.9941
-23.2685
7555.5
.7524
A1
.05543
.00455886
220.9
-.2715
-2.856
.2253
-.1190
-.4910
.00370137
269.4
1.005
-5.873
-.05050
.3326
-.09110
.00360750
275.6
-.02034
2.120
.3034
-.27026
.8326
.00330755
301.6
-1.008
.8193
.09020
.3045
-.1732
.00346462
289.3
.4479
.8952
-.01167
1.077
-.02694
.00308029
320.4
-.3328
6
A2
-.03654
.0000
-.6938
.3203
2.382
-.08275
.02889
.2059
.0000
.3460
-.4015
4.038
.06481
-.1858
.05122
.0000
.7184
.1086
.3703
-.1333
.04560
-.5107
.0000
.3321
1.161
.4269
-.02747
-.1930
.07463
.0000
-.2999
-.07484
-.0756
.01355
-.4757
.01685
.0000
1.906
.4543
A3
.0000
.0000
.0000
-.1230
.0000
.0000
.0000
.0000
.0000
.0000
-.00162
.0000
.0000
.0000
.0000
.0000
.0000
-.05945
.0000
.0000
.0000
.0000
.0000
.0000
-.3777
.0000
.0000
.0000
.0000
.0000
.0000
-.05868
.0000
.0000
.0000
.0000
.0000
.0000
-.1343
Table 1: Polynomial Coefficients for Parameters.
Filter / Ramp
#
FR868N #2
FR868N #3
FR868N #4
Parameter
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
Y
w0
T0
HWHM
a
b
c
A0
49.32
.2958
1.321
.3762
-22.9766
8014.3
.8204
54.17
.05832
.4582
.6422
-22.6085
8510.7
.5817
55.19
.5422
1.420
.4257
-23.2142
9034.3
.6241
59.69
.2376
.9670
.7829
A1
1.742
-.3877
-.9156
1.668
.00286673
350.5
-.3368
1.579
.7525
.8433
.3247
.00265657
375.6
-.1920
-.7459
-.2444
-1.176
-.2522
.00256976
387.2
.2403
2.167
-.01879
.02456
.03750
A2
.4914
.2465
.3666
-.9499
.0000
-.7500
.3815
.2196
-.3625
-.4350
-.1593
.0000
.3706
.4517
1.433
.03545
.4814
.1777
.0000
.8722
-.1230
-.1996
-.00864
-.00477
.02393
A3
.0000
.0000
.0000
.0000
.0000
.0000
-.1057
.0000
.0000
.0000
.0000
.0000
.0000
-.1627
.0000
.0000
.0000
.0000
.0000
.0000
.02829
.0000
.0000
.0000
.0000
4. Calculation of the SYNPHOT LRF Table
We begin by discussing the format of the table, and the way it is utilized by the SYNPHOT
program. After that, we describe in depth how the elements of the table are computed.
4.1 Format and utilization of the SYNPHOT LRF table
SYNPHOT integrates the detected countrate in many wavelength intervals to model the
overall countrate measured by an observer. At each wavelength interval the countrate is
computed as the product of the target countrate, the transmissions of the OTA, WFPC
optics and filter(s), and the CCD response. For standard filters the SYNPHOT filter transmission tables are simply one-dimensional tables listing wavelengths and the transmis-
7
sions at those wavelengths. In contrast, the LRF filters are tunable to many different
wavelengths, and hence its SYNPHOT transmission table is two-dimensional. The dimensions are wavelength setting (i.e. wavelength requested on the Phase II proposal, running
across the rows), and wavelength / transmission as in the ordinary filters (running down
the columns).
Table 2 shows portions of the final SYNPHOT LRF table, and serves to illustrate its forTable 2: Portions of SYNPHOT LRF table.
WAVELENGTH
WAVE#3709.999
WAVE#3714.162
WAVE#3718.329
...
...
1.0000
0.
0.
0.
...
3000.0000
0.
0.
0.
...
...
3001.5000
0.
0.
0.
...
...
3003.0007
0.
0.
0.
...
...
...
...
...
...
...
...
3516.8740
0.
0.
...
...
3518.6331
4.1519e-8
0.
0.
...
...
3520.3921
4.4670e-8
4.6018e-8
0.
...
...
...
...
...
...
...
...
3708.2471
0.069466
0.072649
0.071765
...
...
3710.1011
0.069814
0.074791
0.075431
...
...
3711.9561
0.069380
0.076198
0.078589
...
...
3713.8120
0.068193
0.076787
0.081103
...
...
3715.6689
0.066323
0.076519
0.082843
...
...
3717.5271
0.063870
0.075409
0.083709
...
...
3719.3860
0.060948
0.073525
0.083644
...
...
...
...
...
...
...
...
(2776 rows)
...
...
...
...
...
(998 cols)
mat in more detail. In essence, each column in the table contains the entire transmission
curve for a given central wavelength (WAVE#xxxx) which an observer would specify on
the Phase II proposal. This is equivalent to the transmission curve at a given physical location on the LRF filter glass. The first entry in each row gives the wavelength, and the other
entries give the transmission values at that wavelength. For example, as we go down the
column titled WAVE#3709.999 we see that the transmission peaks at ~0.0698 for WAVELENGTH~3710Å. The next column to the right, which is titled WAVE#3714.162, has a
peak transmission ~0.0768 for WAVLENGTH~3714Å, and so forth.
Observers specify LRF settings in SYNPHOT by placing LRF#xxxx in the PHOTMODE,
8
where xxxx is the central wavelength specified on the Phase II proposal. SYNPHOT will
look amongst the column headers in the table until it finds a pair of WAVE#yyyy entries
which bracket the wavelength specified in the PHOTMODE. Next it looks through the first
column of the table, the WAVELENGTH column, until it finds a pair of values bracketing
the wavelength at which the throughput value is needed. Finally, SYNPHOT does a twodimensional linear interpolation between the four throughput values surrounding the
required WAVE# and WAVELENGTH point.
4.2 Calculation of the SYNPHOT LRF table
We begin with details of the table structure and layout. We have used logarithmic steps in
both WAVE#xxxx and WAVELENGTH when creating the table, since this maximizes the
accuracy for a given table size. The necessity of storing the table in FITS format within the
HST archive limits the number of columns to a maximum of 1000. Hence the WAVE# values are stepped according to
WAVE ( n + 1 ) = 1.001122 × W AVE ( n )
(5)
which leads to 998 steps in WAVE#xxxx between the lowest (WAVE#3710) and highest
(WAVE#9762) settings, when extra values for jumps, etc. are included (as discussed
below). There is no limit to the number of rows, and we have therefore used finer logarithmic steps in WAVELENGTH:
WAVELENGTH ( n + 1 ) = 1.0005 × W AVELENGTH ( n )
(6)
which leads to 2776 rows.
There are also places where sudden jumps are required in the throughput values. These
occur at the start and end of wavelength runs in the TRANS LRF table, where either the
ramp segment, filter, or filter rotation suddenly changes. To handle these we have made
two WAVE#xxxx entries stepped by only 0.001Å at these boundaries. For example, central wavelengths of 3800Å or less are mapped onto filter FR418N, ramp #1, rotation 0˚.
But central wavelengths above 3800Å map onto filter FR418N, ramp #1, rotation -33˚.
Hence we have made two columns in the SYNPHOT LRF table near this wavelength -one at WAVE#3800.000 which ends the first filter setting, and one at WAVE#3800.001
which starts the second setting. Because of this, SYNPHOT LRF users must specify their
wavelength to the nearest 0.01Å. Specifying the wavelength to more than two decimal
places will cause errors at these jumps.
Another important detail is the handling of “vignetted” settings of the LRFs. These are
wavelengths where, for one reason or another, the optimal location on the filter glass cannot be reached (for example, filter areas which lie outside the field imaged by the CCDs).
Instead, these wavelengths are observed at the closest available setting, and hence the
overall transmission curve will be offset slightly in wavelength from the optimal curve for
the requested wavelength. Accordingly, the wavelength requested by the observer will not
be at the peak of the transmission curve, but rather will be offset slightly from the peak,
which will generally imply a small reduction in transmission. Handling of these vignetted
9
regions in the SYNPHOT LRF table is very simple  as with unvignetted wavelengths,
the transmission curve is set to that appropriate for the physical location on the filter glass
used for the observation. As should be apparent, vignetted wavelength ranges in the
TRANS LRF table result in identical adjacent columns in the SYNPHOT LRF table, with
only the WAVE# changing. This occurs is because a single physical location on the filter
glass (with a single transmission curve) is assigned to multiple wavlengths (WAVE#) once
the optimal location on the filter can no longer be reached.
Actual calculation of the table proceeds as follows. One takes each line of the TRANS
LRF table, and steps through the central wavelength range for that line using equation (5),
starting at the low wavelength end (plus 0.001Å), and ending at a the upper wavelength
value. At each wavelength step, one writes a column titled WAVE#xxxx in the LRF table,
where xxxx is the central wavelength. The throughput values in the column are computed
using the following steps:
Set w0 = xxxx, where xxxx is the value in WAVE#xxxx.
Obtain the filter and ramp# for wavelength w0 from the TRANS LRF table.
Compute Y from w0 using equation (4) and appropriate “parameter Y” entries in
Table 2 for that filter and ramp#.
Compute a, b, c, T0, and HWHM from Y using equation (3) and the appropriate
polynomial coefficients in Table 1 for the filter and ramp# specified by the TRANS
LRF table.
Step down the column starting with WAVELENGTH = w = 3000Å and using equation (6), computing a transmission value T at each step using equations (1) and (2),
until the complete transmission curve for that central wavelength setting is
obtained.
Separate rows for 1Å and 999999Å are introduced at the top and bottom of the table, to
prevent erroneous extrapolations at very small and large wavelengths. Errors caused by
quantization and linear interpolation in the table are less than 1% at all wavelengths. Sample transmission curves are shown in the Appendix.
5. Obtaining and Installing the SYNPHOT LRF Table
If your local SYNPHOT tables (e.g., wfpc2 optics, dqe, filter throughputs, etc.) were
updated July 1995 or later, the only new files that will need to be retrieved are the new
LRF table and the new graph and comp tables which point to the new LRF table. Note: if
your local tables predate July 1995, retrieve just the README file from the wfpc2 directory specified below, for detailed instructions on updating all wfpc2-related throughput
tables at once. Retrieval can be done via ftp (commands are given in boldface on the
left):
ftp stsci.edu
login anonymous
binary
give email address as password
set transfer mode to binary
10
cd cdbs/cdbs2/comp
directory for all graph and comp tables
ls hstgraph*
list out graph tables
get hstgraph_960208a.tab or any graph table later than this Feb 8,1996 table1
ls hstcomp*
get hstcomp_960208a.tab or any comp table later than this Feb 8,1996 table1
cd wfpc2
contains all wfpc2-related tables
get wfpc2_lrf_002.tab
new lrf table, old file was *001.tab
quit
That’s it. Once retrieved, the files need to be placed in their respective directories at your
local site: cdbs/comp (graph and comp tables) and cdbs/comp/wfpc2 (lrf table). Two
other useful checks to run, to verify that the LRF component is installed correctly at your
site:
1) The IRAF refdata parameter set should be set, either by executing an unlearn refdata
at the IRAF prompt, or by using epar to examine and alter the parameter set. To point all
the SYNPHOT tasks at the most recent comp and graph table files, refdata should be set as
follows (text shown in boldface is entered by user):
epar refdata
PACKAGE = synphot
TASK = refdata
(area
=
45238.93416) Telescope area in cm^2
(grtbl =
crcomp$hstgraph_*) Instrument graph table
(cmptbl =
crcomp$hstcomp_*) Instrument component table
(mode
=
a)
:quit
Including the literal "_*" in the table name fields will direct the tasks to use the most
recent comp and graph tables in your local cdbs/comp directory (i.e., with subsequent
synphot table updates, you won’t need to change refdata).
2) Check that the new LRF table is being accessed, for example, using the showfiles task:
showfiles "wfpc2,4,a2d7,lrf#3750"
#Throughput table names:
crotacomp$hst_ota_005.tab
crwfpc2comp$wfpc2_optics_003.tab
crwfpc2comp$wfpc2_lrf_002.tab[wave#]
crwfpc2comp$wfpc2_dqewfc4_002.tab
crwfpc2comp$wfpc2_a2d7wf4_002.tab
<--should see wfpc2_lrf_002.tab
Note that specifying the filtername (e.g. FR418N) in the OBSMODE is not necessary; the
pipeline (CALWP2) currently includes the filter in the PHOTMODE (keyword in the
image header, equivalent to obsmode in SYNPHOT parameter sets), but it does not
1. Since
all instruments use these hstgraph* and hstcomp* tables, many synphot
updates do not affect wfpc2 throughput data (e.g., the later Feb 9, 1996 update did
not include wfpc2-related changes).
11
include any additional throughput component. The showfiles command will report a message to this effect:
showfiles "wfpc2,4,a2d7,fr418n,lrf#3750"
WARNING Instrument mode keywords not used: fr418n
#Throughput table names:
crotacomp$hst_ota_005.tab
crwfpc2comp$wfpc2_optics_003.tab
crwfpc2comp$wfpc2_lrf_002.tab[wave#]
crwfpc2comp$wfpc2_dqewfc4_002.tab
crwfpc2comp$wfpc2_a2d7wf4_002.tab
6. How to Perform LRF Photometric Calibration with SYNPHOT
Prior to making photometric measurements, the LRF images should be flat fielded with a
narrow band filter flat observed at a nearby wavelength. As of this writing, this is not automatically done during the “pipeline” calibration. For example, near 5000Å one would use
a flat observed through the F502N filter. The WFPC2 WWW pages at address http://
www.stsci.edu/ftp/instrument_news/WFPC2/wfpc2_top.html should be consulted
for the latest flat field reference files. A suitable flat should be obtained from the HST data
archive, and then multiplied into the LRF images. Table 3 lists the current narrow band
flats, but observers should also check the WWW pages for updated versions of these flats.
Wide band filter F791W is included to avoid a large gap at those wavelengths.
Table 3: Recommended Flat Fields
λ(Å)
FILTER
NAME
FLAT FIELD
FILE
MODE
3750
F375N
e3809349u
full
3900
F390N
e380934eu
full
4370
F437N
e380934ju
full
4690
F469N
e380934su
full
4870
F487N
e3809351u
full
5020
F502N
e3809354u
full
5880
F588N
e3914337u
full
6310
F631N
e391433gu
full
6560
F656N
e391433ju
full
6580
F658N
e391433lu
full
6730
F673N
e391433ru
full
7910
F791W
e391434bu
full
9530
F953N
e391434mu
full
We strongly recommend using narrow band filter flats, since these are insensitive to the
12
spectral variations in the flat field light source. If instead, one were to use a flat taken
through the LRF, a very accurate spectrum would be needed for the light source so that its
spectral variations could be corrected during photometric calibration. Using a narrow band
flat eliminates these uncertainties. In addition, the F418N LRF filter also contains a number of pinholes; while these usually have no impact on science images, they seriously corrupt flats taken through that filter. For these reasons we recommend flat fielding with one
of the narrow band flats listed in Table 3. Tests indicate they should give results accurate to
a few percent (2% to 3%).
After flattening the data, count levels should be measured for the target in the usual manner. Unlike the normal filters, the target location for the LRFs depends on the wavelength
specified in the Phase II proposal, and may not be obvious if the target is faint. The target
locations (CCD and pixel numbers) can be estimated using Tables 3.7 and 3.8 in the
WFPC2 Handbook (V.3.0 or later), or the LRF Calculator Tool on the WFPC2 WWW
pages.
The conversion from target counts to target flux is performed using SYNPHOT. The CALCPHOT task in SYNPHOT will compute the expected count rate for a variety of model
targets, and hence provides the necessary scaling from counts to flux. There are three
essential ingredients for this calculation: an OBSMODE which describes the instrumental
parameters, a WAVETAB which lists the wavelength intervals used for the calculation, and
a SPECTRUM containing a model of the target spectrum.
Observers specify LRF settings in the OBSMODE by including LRF#xxxx, where xxxx is
the central wavelength specified on the Phase II proposal. It is imperative that the number
from the Phase II proposal be used, since this selects the filter transmission curve. The
value from the Phase II must be used, regardless of whether the LRF setting is vignetted,
or whether the emission line was really at a different wavelength than specified on the
Phase II. Also, as discussed in Section 4, the central wavelength must be specified to two
decimal places or less (e.g. 5069.34 or 5069, but not as 5069.343). The total HST and
WFPC2 efficiencies should also be included by specifying wfpc2,3. The “3” implies
CCD WF3, but the identical results will be obtained regardless of which CCD is specified.
Also, a2d7 or a2d15 should be given depending on whether the gain was set to 7 or 14,
respectively, during the observation. Hence the OBSMODE would be specified as (e.g.):
obsmode = wfpc2,3,lrf#5007,atd7
where the central wavelength is 5007Å in this example. The LRF bandpass, either with or
without the OTA+WFPC2+CCD efficiencies, can be plotted with the SYNPHOT
PLBAND task as shown in the Appendix.
The default WAVETAB or wavelength table used by SYNPHOT can have steps which are
too coarse to accurately integrate a narrow filter, such as the LRFs. This can be remedied
by using the GENWAVE command in SYNPHOT to custom-design a wavelength table. In
general, 1Å steps should be more than sufficient for the LRFs, although the wavelength
steps must also be fine enough to well-sample any narrow lines in the target spectrum. The
name of this table is then given as the WAVETAB parameter in the various SYNPHOT
tasks.
13
Here is an example of running GENWAVE. First we start IRAF and load the SYNPHOT
package, and then we run GENWAVE:
cl
stsdas
hst_calib
synphot
genwave wavelen.tab 1000. 12000. 1.
This produces a table from 1000Å to 12000Å in 1Å steps.
We next discuss the specification of model spectra for emission line, stellar, and powerlaw
targets.
6.1 Emission line spectra
Modeling of spectral lines will generally require use of the either the BOX or GAUSS
passband functions to generate a model line profile. While the SYNPHOT User’s Guide
describes these as part of the OBSMODE, or instrumental parameters, they can also be
used to specify a target spectrum. For example, we can multiply a box passband by a constant spectrum to obtain a box-shaped emission line:
spectrum = box( 5007, 10) * unit(1.0e-15,flam)
which describes a box emission line 10Å wide centered at 5007Å having peak flux of
1.0x10-15 erg sec-1 cm-2 Å-1, and hence a total flux of 1.0x10-14 erg sec-1 cm-2. Alternatively, an emission line can be modeled as a Gaussian function:
spectrum = gauss( 5007, 10) * unit(1.0e-15,flam)
which describes a Gaussian line profile with 10Å FWHM centered at 5007Å and a peak
flux density 1.0 x10 -15 erg sec-1 cm-2 Å-1. The total flux in such a Gaussian spectrum will
be 1.054(FWHM)(peak flux density) or 1.05 x10 -14 erg sec-1 cm-2 for this example. Spectra can also be added, if more than one emission line is needed. For example:
spectrum = gauss(5007,10)*unit(1.e-18,flam) + gauss(5020,10)*
unit(2.e-18,flam)
where all the text is place on a single line.
The actual photometric calculations are performed using CALCPHOT. Assuming IRAF is
running, the SYNPHOT package is loaded, and that GENWAVE was run as described
above, we then run CALCPHOT. Typing epar calcphot brings up a fill-out form as
shown below; values are typed in, and <return> is used to go to the next line. After entering the various parameters, we then type :go to start the calculation.
Here is an example of a CALCPHOT run for an LRF central wavelength of 5007Å, and a
model Gaussian spectrum centered at 5010Å having a FWHM of 5Å and total flux of
14
1.05x10 -13 erg sec-1 cm-2:
sy>epar calcphot
Image Reduction and Analysis Facility
PACKAGE = synphot
TASK = calcphot
obsmode = wfpc2,3,lrf#5007,a2d7 Instrument observation mode
spectrum= gauss(5010,5)*unit(2.0e-14,flam) Synthetic spectrum
form
=
counts Form for output data
(vzero =
) List of values for variable zero
(output =
none) Output table name
(append =
no) Append to existing table?
(result =
0.) Result of synphot calculation
(wavetab=
wavelen.tab) Wavelength table name
(refdata=
) Reference data
(mode
=
a)
:go
Mode = band(wfpc2,3,lrf#5007,a2d7)
Pivot
Equiv Gaussian
Wavelength
FWHM
5007.47
64.40982
band(wfpc2,3,lrf#5007,a2d7)
Spectrum: gauss(5010,5)*unit(2.0e-14,flam)
VZERO
(COUNTS s^-1 hstarea^-1)
0.
4.52047
The results indicate that the target flux 1.05x10 -13 erg sec-1 cm-2 will produce 4.5 DN s-1.
This can then be scaled to the actual countrate in the observed image, to obtain the target
flux.
More complex spectra may require more characters than the fill-out form allows. In such
cases the spectrum can be specified in a text file, and the name of the text file (proceeded
by “@”) is given for the spectrum in CALCPHOT. For example, the following spectrum
could be entered into a text file called spectrum.dat with any editor:
gauss(5007,15)*unit(1.e-16,flam)+gauss(5028,8)*unit(2.e-16,flam)
and then CALCPHOT is run on this spectrum:
sy>epar calcphot
Image Reduction and Analysis Facility
PACKAGE = synphot
TASK = calcphot
obsmode = wfpc2,3,lrf#5007,a2d7
15
Instrument observation mode
spectrum=
form
=
(vzero =
(output =
(append =
(result =
(wavetab=
(refdata=
(mode
=
@spectrum.dat
counts
)
none)
no)
0.)
wavelen.tab)
)
a)
Synthetic spectrum to calculate
Form for output data
List of values for variable zero
Output table name
Append to existing table?
Result of synphot calculation
Wavelength table name
Reference data
:go
Mode = band(wfpc2,3,lrf#5007,a2d7)
Pivot
Equiv Gaussian
Wavelength
FWHM
5007.47
64.40982
band(wfpc2,3,lrf#5007,a2d7)
Spectrum: gauss(5007,15) * unit(1.0e-16,flam) + gauss(5028,8) *
unit(2.0e-16,flam)
VZERO
(COUNTS s^-1 hstarea^-1)
0.
0.117225
The results indicate that the model spectrum will produce 0.12 DN s-1. The input spectrum
can be plotted using the PLSPEC task. Here we plot the input spectrum alone by leaving
the OBSMODE blank, and request that the plot be in units of Fλ or “flam” (i.e. erg sec-1
cm-2 Å-1). The plot limits are set to 4900Å to 5100Å:
sy>epar plspec
Image Reduction and Analysis Facility
PACKAGE = synphot
TASK = plspec
obsmode =
spectrum=
form
=
(vzero =
(spfile =
(pfile =
(errtyp =
(left
=
(right =
(bottom =
(top
=
(append =
(ltype =
(device =
(wavetab=
(refdata=
(mode
=
@spectrum.dat
flam
)
none)
none)
n)
4900.)
5100.)
INDEF)
INDEF)
no)
solid)
stdgraph)
wavelen.tab)
)
a)
:go
16
Observation mode or @list
Synthetic spectrum or @list
Form of output graph
List of values for variable zero
Spectrophotometry data
Photometry data
n[one] p[oint] c[ont] b[in]...
x value for left side of graph
x value for right side
y value for bottom
y value for top
Append to existing plot?
Line type: clear,solid,dashed,...
Graphics device
Wavelength table name
Reference data
The resulting model target spectrum is shown in Figure 1.
Figure 1. Example of PLSPEC output for emission lines.
FLAM (erg cm-2 s-1 Å-1)
2.00E-16
1.50E-16
1.00E-16
5.00E-17
0
4900
4950
5000
Wavelength (Å)
5050
5100
6.2 Stellar spectra
Stellar spectra can be modeled using catalogs of either observed or synthetic spectra incorporated in SYNPHOT. Appendix B in the SYNPHOT User’s Guide lists the available spectra. For example, if we needed a model for an M6 III star with V=16.6, we could use star
#166 in the Bruzual-Perrson-Gunn-Stryker Spectral Atlas, which is listed in Table B.5 of
the User’s Guide as bpgs_166, and re-normalize it to V=16.6 as follows:
spectrum = rn(crgridbpgs$bpgs_166,band(v),16.6,vegamag)
Using this as an input for CALCPHOT, we have:
sy>epar calcphot
Image Reduction and Analysis Facility
PACKAGE = synphot
TASK = calcphot
obsmode = wfpc2,3,lrf#5007,a2d7 Instrument observation mode
spectrum= rn(crgridbpgs$bpgs_166,band(v),16.6,vegamag)
17
form
=
(vzero =
(output =
(append =
(result =
(wavetab=
(refdata=
(mode
=
counts
)
none)
no)
0.)
wavelen.tab)
)
a)
Form for output data
List of values for variable zero
Output table name
Append to existing table?
Result of synphot calculation...
Wavelength table name
Reference data
:go
Mode = band(wfpc2,3,lrf#5007,a2d7)
Pivot
Equiv Gaussian
Wavelength
FWHM
5007.47
64.40982
band(wfpc2,3,lrf#5007,a2d7)
Spectrum: rn(crgridbpgs$bpgs_166,band(v),16.6,vegamag)
VZERO
(COUNTS s^-1 hstarea^-1)
0.
0.828713
Hence this star observed with the LRFs at a central wavelength of 5007Å results in 0.83
DN s-1. The input spectrum may be plotted as above using PLSPEC, with the result shown
in Figure 2.
Figure 2. Example of PLSPEC output for M6 III star.
FLAM (erg cm-2 s-1 Å-1)
1.25E-15
1.00E-15
7.50E-16
5.00E-16
2.50E-16
4900
4950
5000
Wavelength (Å)
18
5050
5100
6.3 Power law spectra
Power law spectra can be modeled using the pl function. For example,
spectrum = 2.3*pl(5500,0.7)
specifies a power law spectrum
ν
λ 0.7
 –0.7
= 0.23  -------------------------f ν = 0.23  ------------
14
 5500

5.45 × 10
where λ is in Å, and f ν is in Janskys. This may also be re-written in terms of frequency ν
in Hz, as shown. This can also be re-normalized to a standard magnitude:
spectrum = rn(pl(5500,0.7),band(v),16.3,vegamag)
where the power law is now specified to have V=16.3. These spectra can be used in CALCPHOT in the same manner as shown above for emission lines and stellar spectra, and
similarly be plotted with PLSPEC.
In closing we recommend that LRF observers consult the WFPC2 LRF calibration web
page (see references) for any developments and updates regarding LRF calibration.
7. References
Biretta, J. A., Ritchie, C. E., Baggett, S., and MacKenty, J. W. 1996a, “Wavelength / Aperture Calibration of
the WFPC2 Linear Ramp Filters.” WFPC2 Instrument Science Report 96-05.
Biretta, J. A., et al., 1996b, WFPC2 Instrument Handbook, Version 4.0.
Bushouse, H., et al. 1992, “SYNPHOT User’s Guide.” 3rd ed.
Evans, R. 1992, “WFPC-2 Ramp Filter Predictors” JPL Interoffice Memorandum DM# 2031 (Dec. 30,
1992).
WWW page: http://www.stsci.edu/ftp/instrument_news/WFPC2/wfpc2_top.html
19
8. Appendix: Sample LRF + System Efficiency Curves
The curves below were generated in SYNPHOT using the following commands. Assuming the new SYNPHOT LRF table has been obtained and installed as described in Section
5, we start IRAF and load the SYNPHOT package. We then generate a finely stepped
wavelength table with GENWAVE, and compute the throughput plots with PLBAND.
Entering epar plband produces a fill-out form, which is completed as below by entering
the parameters and then hitting <return>. After the fill-out form is completed, typing :go
will run PLBAND and produce a plot on the screen. Text shown in boldface is entered by
the user:
cl
stsdas
hst_calib
synphot
genwave wavelen.tab 1000. 12000. 1.
epar plband
Image Reduction and Analysis Facility
PACKAGE = synphot
TASK = plband
obsmode =
(left
=
(right =
(bottom =
(top
=
(normali=
(ylog
=
(append =
(ltype =
(device =
(wavetab=
(refdata=
(mode
=
wfpc2,3,lrf#5000
4900.)
5100.)
INDEF)
INDEF)
no)
no)
no)
solid)
stdgraph)
wavelen.tab)
)
a)
Instrument mode or @list
x value for left side of plot
x value for right side of plot
y value for bottom of plot
y value for top of plot
Normalization all curves to 1?
Take log of y values?
Append to existing plot?
Line type: clear,solid,dashed,dotted,...
Graphics device
Wavelength table name
Reference data
:go
In this example the central wavelength has been set to 5000Å (i.e. the central wavelength
specified on the Phase II proposal), and the plot runs from 4900Å to 5100Å. A hard copy
can be generated by entering device=stdplot, and then typing gflush after PLBAND is
done. Here the OTA+WFPC2 optics and CCD are included in the calculation; to plot the
transmission of the LRF in isolation, omit “wfpc2,3” from the OBSMODE.
The plots in Figures A.1. through A.11. give the total efficiency for the OTA+WFPC2
optics, LRF, and CCDs together at central wavelengths from 3710Å to 9762Å with steps
of 100Å between plots. Finer 50Å steps are used at the ends of the range.
20
.0035
.01
OBSMODE = wfpc2,3,lrf#3800
.015
OBSMODE = wfpc2,3,lrf#3900
.0015
PASSBAND
.01
.002
PASSBAND
PASSBAND
.0125
.008
.0025
.006
.0075
.004
.005
1.00E-3
.002
.0025
5.00E-4
0
0
3650
3675
3700
3725
3750
WAVELENGTH (A)
3775
3725
0
3750
3775
3800
3825
WAVELENGTH (A)
3850
3875
3825
3850
3875
3900
3925
WAVELENGTH (A)
3975
4050
4075
21
3950
.007
OBSMODE = wfpc2,3,lrf#3750
.015
OBSMODE = wfpc2,3,lrf#3850
.006
OBSMODE = wfpc2,3,lrf#4000
.0125
.0125
.005
.01
.003
PASSBAND
PASSBAND
PASSBAND
.01
.004
.0075
.0075
.005
.005
.002
.0025
.001
0
3675
.0025
0
0
3700
3725
3750
3775
WAVELENGTH (A)
3800
3825
3775
3800
3825
3850
3875
WAVELENGTH (A)
3900
3925
3925
3950
3975
4000
4025
WAVELENGTH (A)
Figure A.1. LRF + System Efficiency for 3710Å through 4000Å.
OBSMODE = wfpc2,3,lrf#3710
.003
OBSMODE = wfpc2,3,lrf#4100
.02
OBSMODE = wfpc2,3,lrf#4300
OBSMODE = wfpc2,3,lrf#4500
.02
.0125
.015
.015
.0075
PASSBAND
PASSBAND
PASSBAND
.01
.01
.01
.005
.005
.005
.0025
0
4025
0
4050
4075
4100
4125
WAVELENGTH (A)
4150
4175
4225
0
4250
4275
4300
4325
WAVELENGTH (A)
4350
4375
4425
22
.025
.0125
OBSMODE = wfpc2,3,lrf#4200
.02
.0075
.015
4475
4500
4525
WAVELENGTH (A)
4550
4575
4650
4675
.03
OBSMODE = wfpc2,3,lrf#4400
.01
4450
OBSMODE = wfpc2,3,lrf#4600
.025
.005
PASSBAND
PASSBAND
PASSBAND
.02
.01
.015
.01
.005
.0025
.005
0
0
4125
4150
4175
4200
4225
WAVELENGTH (A)
4250
4275
4325
0
4350
4375
4400
4425
WAVELENGTH (A)
4450
4475
4525
4550
4575
4600
4625
WAVELENGTH (A)
Figure A.2. LRF + System Efficiency for 4100Å through 4600Å.
.015
.035
.05
OBSMODE = wfpc2,3,lrf#4900
OBSMODE = wfpc2,3,lrf#5100
.03
.03
.04
.02
.02
PASSBAND
PASSBAND
PASSBAND
.025
.015
.03
.02
.01
.01
.01
.005
0
4625
0
0
4650
4675
4700
4725
WAVELENGTH (A)
4750
4775
4825
4850
4875
4900
4925
WAVELENGTH (A)
4950
4975
5000
5050
5100
5150
WAVELENGTH (A)
5200
23
.03
OBSMODE = wfpc2,3,lrf#4800
.025
OBSMODE = wfpc2,3,lrf#5000
.04
OBSMODE = wfpc2,3,lrf#5200
.025
.02
.03
.015
PASSBAND
PASSBAND
PASSBAND
.02
.015
.02
.01
.01
.01
.005
.005
0
4725
0
0
4750
4775
4800
4825
WAVELENGTH (A)
4850
4875
4900
4950
5000
5050
WAVELENGTH (A)
5100
5100
5150
5200
5250
WAVELENGTH (A)
5300
Figure A.3. LRF + System Efficiency for 4700Å through 5200Å.
OBSMODE = wfpc2,3,lrf#4700
.04
.06
OBSMODE = wfpc2,3,lrf#5500
OBSMODE = wfpc2,3,lrf#5700
.05
.05
.03
.04
.02
PASSBAND
PASSBAND
PASSBAND
.04
.03
.03
.02
.02
.01
.01
.01
0
0
5200
5250
5300
5350
WAVELENGTH (A)
5400
5400
0
5450
5500
5550
WAVELENGTH (A)
5600
5600
5650
5700
5750
WAVELENGTH (A)
5800
24
.06
.05
OBSMODE = wfpc2,3,lrf#5400
.06
OBSMODE = wfpc2,3,lrf#5600
OBSMODE = wfpc2,3,lrf#5800
.05
.05
.04
.04
.03
PASSBAND
PASSBAND
PASSBAND
.04
.03
.03
.02
.01
0
5300
5350
5400
5450
WAVELENGTH (A)
5500
.02
.02
.01
.01
0
0
5500
5550
5600
5650
WAVELENGTH (A)
5700
5700
5750
5800
5850
WAVELENGTH (A)
5900
Figure A.4. LRF + System Efficiency for 5300Å through 5800Å
OBSMODE = wfpc2,3,lrf#5300
.06
OBSMODE = wfpc2,3,lrf#6100
.1
OBSMODE = wfpc2,3,lrf#6300
.08
.05
.08
.03
PASSBAND
.06
PASSBAND
PASSBAND
.04
.04
.06
.04
.02
.02
.02
.01
0
5800
0
0
5850
5900
5950
WAVELENGTH (A)
6000
6000
6050
6100
6150
WAVELENGTH (A)
6200
6200
6250
6300
6350
WAVELENGTH (A)
6400
25
.1
OBSMODE = wfpc2,3,lrf#6000
OBSMODE = wfpc2,3,lrf#6200
OBSMODE = wfpc2,3,lrf#6400
.08
.04
.08
.06
.02
PASSBAND
PASSBAND
PASSBAND
.03
.06
.04
.04
.02
.01
.02
0
5900
0
0
5950
6000
6050
WAVELENGTH (A)
6100
6100
6150
6200
6250
WAVELENGTH (A)
6300
6300
6350
6400
6450
WAVELENGTH (A)
6500
Figure A.5. LRF + System Efficiency for 5900Å through 6400Å.
OBSMODE = wfpc2,3,lrf#5900
.125
OBSMODE = wfpc2,3,lrf#6700
OBSMODE = wfpc2,3,lrf#6900
.1
.08
.1
.08
.06
.05
PASSBAND
PASSBAND
PASSBAND
.075
.06
.04
.04
.02
.025
.02
0
0
6400
6450
6500
6550
WAVELENGTH (A)
6600
0
6600
6650
6700
6750
WAVELENGTH (A)
6800
6800
6850
6900
6950
WAVELENGTH (A)
7000
26
.1
OBSMODE = wfpc2,3,lrf#6600
OBSMODE = wfpc2,3,lrf#6800
OBSMODE = wfpc2,3,lrf#7000
.08
.1
.08
.06
.05
.06
PASSBAND
PASSBAND
PASSBAND
.075
.04
.04
.025
.02
.02
0
0
0
6500
6550
6600
6650
WAVELENGTH (A)
6700
6700
6750
6800
6850
WAVELENGTH (A)
6900
6900
6950
7000
7050
WAVELENGTH (A)
7100
Figure A.6. LRF + System Efficiency for 6500Å through 7000Å.
OBSMODE = wfpc2,3,lrf#6500
.08
OBSMODE = wfpc2,3,lrf#7300
OBSMODE = wfpc2,3,lrf#7500
.08
.06
.06
.04
PASSBAND
PASSBAND
PASSBAND
.06
.04
.02
.04
.02
.02
0
0
0
7000
7050
7100
7150
WAVELENGTH (A)
7200
7200
7250
7300
7350
WAVELENGTH (A)
7400
7400
7450
7500
7550
WAVELENGTH (A)
7600
27
.08
.08
OBSMODE = wfpc2,3,lrf#7200
OBSMODE = wfpc2,3,lrf#7400
OBSMODE = wfpc2,3,lrf#7600
.06
PASSBAND
PASSBAND
.06
.04
.04
.02
.02
0
0
7100
7150
7200
7250
WAVELENGTH (A)
7300
PASSBAND
.06
.04
.02
0
7300
7350
7400
7450
WAVELENGTH (A)
7500
7500
7550
7600
7650
WAVELENGTH (A)
7700
Figure A.7. LRF + System Efficiency for 7100Å through 7600Å.
OBSMODE = wfpc2,3,lrf#7100
OBSMODE = wfpc2,3,lrf#7900
.04
.04
.03
.05
.04
PASSBAND
.05
PASSBAND
.05
OBSMODE = wfpc2,3,lrf#8100
.03
.03
.02
.02
.02
.01
.01
.01
0
0
0
7600
7650
7700
7750
WAVELENGTH (A)
7800
7800
7850
7900
7950
WAVELENGTH (A)
8000
7950
8000
8050 8100 8150 8200
WAVELENGTH (A)
8250
28
.06
.06
OBSMODE = wfpc2,3,lrf#7800
OBSMODE = wfpc2,3,lrf#8000
.05
.05
.04
.04
.03
.03
PASSBAND
PASSBAND
PASSBAND
OBSMODE = wfpc2,3,lrf#8200
.04
.03
.02
.02
.01
.01
.02
.01
0
0
7700
7750
7800
7850
WAVELENGTH (A)
7900
0
7850
7900
7950 8000 8050 8100
WAVELENGTH (A)
8150
8050
8100
8150 8200 8250 8300
WAVELENGTH (A)
8350
Figure A.8. LRF + System Efficiency for 7700Å through 8200Å.
OBSMODE = wfpc2,3,lrf#7700
PASSBAND
.06
.06
.06
.04
OBSMODE = wfpc2,3,lrf#8300
.025
PASSBAND
PASSBAND
.02
OBSMODE = wfpc2,3,lrf#8700
.02
.03
.03
PASSBAND
OBSMODE = wfpc2,3,lrf#8500
.02
.015
.01
.01
.01
.005
0
0
8150
8200
8250 8300 8350 8400
WAVELENGTH (A)
8450
0
8350
8400
8450 8500 8550 8600
WAVELENGTH (A)
8650
8550
8600
8650 8700 8750 8800
WAVELENGTH (A)
8850
29
.04
.04
OBSMODE = wfpc2,3,lrf#8400
OBSMODE = wfpc2,3,lrf#8600
OBSMODE = wfpc2,3,lrf#8800
.025
.03
.02
.02
PASSBAND
PASSBAND
PASSBAND
.03
.02
.015
.01
.01
.01
.005
0
0
0
8250
8300
8350 8400 8450 8500
WAVELENGTH (A)
8550
8450
8500
8550 8600 8650 8700
WAVELENGTH (A)
8750
8650
8700
8750 8800 8850 8900
WAVELENGTH (A)
8950
Figure A.9. LRF + System Efficiency for 8300Å through 8800Å.
.04
OBSMODE = wfpc2,3,lrf#9100
OBSMODE = wfpc2,3,lrf#9300
.025
.02
.015
.015
PASSBAND
PASSBAND
PASSBAND
.02
.015
.01
.01
.01
.005
.005
.005
0
0
8750
8800
8850 8900 8950 9000
WAVELENGTH (A)
9050
0
8950
9000
9050 9100 9150 9200
WAVELENGTH (A)
9250
9150
9200
9250 9300 9350 9400
WAVELENGTH (A)
9450
30
OBSMODE = wfpc2,3,lrf#9000
OBSMODE = wfpc2,3,lrf#9200
OBSMODE = wfpc2,3,lrf#9400
.025
.02
.015
.02
.015
PASSBAND
PASSBAND
PASSBAND
.015
.01
.01
.01
.005
.005
.005
0
0
8850
8900
8950 9000 9050 9100
WAVELENGTH (A)
9150
0
9050
9100
9150 9200 9250 9300
WAVELENGTH (A)
9350
9250
9300
9350 9400 9450 9500
WAVELENGTH (A)
9550
Figure A.10. LRF + System Efficiency for 8900Å through 9400Å.
.02
.025
OBSMODE = wfpc2,3,lrf#8900
OBSMODE = wfpc2,3,lrf#9500
.01
OBSMODE = wfpc2,3,lrf#9650
.0125
OBSMODE = wfpc2,3,lrf#9750
.01
.008
.0075
.0075
PASSBAND
PASSBAND
PASSBAND
.01
.005
.006
.004
.005
.0025
.002
.0025
0
0
9350
9400
9450 9500 9550 9600
WAVELENGTH (A)
9650
0
9500
9550
9600 9650 9700 9750
WAVELENGTH (A)
9800
9600
9650
9700 9750 9800 9850
WAVELENGTH (A)
9900
31
.01
.0125
OBSMODE = wfpc2,3,lrf#9600
OBSMODE = wfpc2,3,lrf#9700
OBSMODE = wfpc2,3,lrf#9762
.01
.008
.01
.0075
PASSBAND
PASSBAND
PASSBAND
.0075
.005
.006
.004
.005
.0025
.002
.0025
0
0
9450
9500
9550 9600 9650 9700
WAVELENGTH (A)
9750
0
9550
9600
9650 9700 9750 9800
WAVELENGTH (A)
9850
9600
9650
9700 9750 9800 9850
WAVELENGTH (A)
9900
Figure A.11. LRF + System Efficiency for 9500Å through 9762Å.
.0125
.015